AMD has introduced support for Radeon AI PRO R9700 and PRO W7900 GPUs as part of its enterprise AI reference stack. The technical preview includes the AMD GPU Operator and AMD Inference Microservices (AIMs), enabling users to deploy and manage AI workloads on AMD hardware. These tools provide standardized, portable inference microservices for serving AI models, abstracting complexities through intelligent orchestration and OpenAI-compatible APIs. The stack also includes AMD AI Workbench and AMD Resource Manager to manage models, workflows, and resources. Source: amd
The AMD Solution Blueprints Catalog offers ready-to-deploy applications, such as the MRI Analysis Tool, which combines computer vision, deep learning, and an AIM to generate AI reports from MRI scans. The tool supports multi-format input, preprocessing, tissue segmentation, anomaly detection, and quantitative measurements, with an LLM-assisted report and interactive Q&A. Deployment involves Helm charts and Kubernetes, with the MRI Analysis Tool serving as a concrete example of the stack’s capabilities. Source: amd
The deployment process includes identifying the chart name, defining parameters, and applying manifests to Kubernetes clusters. The MRI Analysis Tool runs as a single container alongside the AIM, with the Qwen3 VL 8B Instruct model as the default LLM. Users can interact with the workflow through a web interface, and the solution blueprint can be deployed on both AMD Instinct and Radeon hardware. Source: amd
